Regulation of Gli Activity by All-Trans Retinoic Acid in Mouse Keratinocytes1

[CANCER RESEARCH 60, 5386–5389, October 1, 2000] Advances in Brief

Regulation of Gli Activity by All-trans Retinoic Acid in Mouse Keratinocytes1

Philippe Goyette, Deborah Allan, Pascal Peschard,2 Chang Feng Chen, Wei Wang, and David Lohnes3 Department of Molecular Biology, Universite´de Montreál [D. L.]; Division of Experimental Medicine, McGill University [D. A., C. F. C., D. L.]; and the Institut de Recherches Cliniques de Montreál [P. G., D. A., P. P., C. F. C., W. W., D. L.], Montreál, Que´bec, H2W 1R7 Canada

Abstract human malignancies, including diverse epithelial cancers or precan- cerous lesions (10–12). Sonic hedgehog (Shh) signaling is essential for many normal develop- The retinoid signal is transduced by two families of nuclear recep- mental processes. The Shh signal is interpreted by the Gli transcription tors, the RARs (RAR␣, RAR␤, RAR␥, and their isoforms) and the factors. Elevated Gli-1 expression has been associated with several neo- RXRs [RXR␣, RXR␤, RXR␥, and their isoforms (13)]. RARs func- plasms, including basal cell carcinoma. All-trans retinoic acid (RA) has tion as ligand-inducible transcription regulators by binding, together strong effects on epidermal growth and differentiation and has been used with a RXR partner, to cis-acting RA response elements. RARs can be for the treatment of various epithelial disorders. In this report, we show that RA can inhibit Gli activity in immortalized murine keratinocytes in activated by both RA and its stereoisomer, 9-cis RA, whereas RXRs a RA receptor-specific manner. This inhibition may occur, at least in part, are activated only by 9-cis RA. RXRs are also heterodimeric partners through sequestration of the transcriptional coactivator cyclic AMP- for other nuclear receptors, including thyroid hormone, vitamin D, responsive element-binding protein-binding protein and suggests a novel peroxisome proliferator activated receptor, and several orphan receptors. effect of retinoid excess on Shh signaling. Normal epidermis expresses RAR␥ and RAR␣ as well as RXR␣ and RXR␤, with RAR␥ and RXR␣ believed to be the predominant Introduction heterodimer (14). Although RAR target genes in skin are largely unknown, retinoid signaling is essential for epidermal development BCC4 is the most common type of skin cancer in Caucasian and maintenance. In addition to this, epidemiological studies and populations. Studies of sporadic and familial BCC have indicated that clinical trials also suggest that retinoids may reduce the incidence of a majority of these tumors show aberrant activation of the Shh non-melanoma skin cancer (15). pathway (1–3). Inactivating mutations of Ptc, a putative receptor and RARs can also function in a ligand-dependent manner to inhibit negative regulator of Shh signaling, have been identified in familial AP-1 activity, and it has been suggested that the antitumorigenic BCCs and sporadic BCCs, whereas activating mutations in Smo, a effects of retinoids may occur through this mechanism (16). The basis positive regulator of the Shh pathway and the target of Ptc repression, for this trans-repression is believed to be due in part to competition have also been observed in sporadic BCCs (1). These mutations lead for limited amounts of ancillary factors, such as CBP or p300, which to elevated expression of Gli-1 (and Gli-2 in some cases), which is are common to both AP-1 and RAR transcription complexes (17). believed to play a causative role in BCC formation. Ptc mutation is the Because both the Drosophila homologue of the Gli transcription underlying basis of basal cell nevus syndrome, which is also associ- factor family, cubitus interruptus, and vertebrate Gli-3 have been ated with familial BCC as well as medulloblastoma and meningioma shown to use CBP as a coactivator (18, 19), Gli signaling is a potential (4). These tumors are also seen in Ptc null mutant mice, and both target for RA-induced trans-repression. To test this hypothesis, we tumor types exhibit elevated expression of Gli-1 (5). Overexpression investigated the effects of RA treatment and RAR expression on Gli of Gli-1 in frog skin causes tumors (2), and its induction in murine signaling activity in transformed mouse keratinocytes. These cell lines express Gli-2 and Gli-3, whereas Gli-1 is not detectable. We found epithelium (via a Shh transgene) results in BCC (3). The possibility that RA excess inhibited the activity of a Gli reporter and that this that Gli family members could serve as proto-oncogenes is also inhibition could be partially reversed by exogenous CBP or p300. consistent with discovery of Gli-1 amplification in human glioma and Attenuation of Gli activity also paralleled Gli-2 levels, and overex- sarcoma (6, 7). pression of Gli-2 partially restored activity. Inhibition was not ob- Vitamin A derivatives (retinoids) play central roles in embryonic served in RAR␣␥ null keratinocytes and was minimal in RAR␥ mutant development and in the maintenance of various tissues in the adult (8, cells, demonstrating that this effect was mediated principally via 9). Retinoids also exhibit potent antitumorigenic properties in many RAR␥. model systems and show potential for the treatment of a number of Materials and Methods Received 12/28/99; accepted 8/14/00. The costs of publication of this article were defrayed in part by the payment of page Cell Culture and Reagents. The RAR null keratinocyte cell lines were charges. This article must therefore be hereby marked advertisement in accordance with established from primary keratinocytes by immortalization with dominant 18 U.S.C. Section 1734 solely to indicate this fact. negative p53 as described previously (20). The cells were grown in SMEM 1 Supported by the National Cancer Institute of Canada with funds from the Canadian supplemented with 10% chelex-treated FCS, (the calcium concentration was Cancer Society. P. G. was supported by a fellowship from the MRC, C. F. C. was ␮ ␮ supported by a scholarship from the Cancer Research, and D. A. was supported by a adjusted to 0.5 mM after treatment), 5 g/ml insulin, 0.5 M hydrocortisone, ϫ Ϫ11 ␮ scholarship from the Natural Sciences and Engineering Council of Canada. D. L. is a 1.5 mM MgCl2, 1.2 10 M cholera toxin, 24 g/ml adenine, 10 ng/ml chercheur boursier (Junior 2) of the Fonds de la Recherches en Sante´de Que´bec. epidermal growth factor, and 10 ␮g/ml gentimycin at 37°C in 5% CO in air 2 2 Present address: Molecular Oncology Group, Royal Victoria Hospital, McGill Uni- in a humidified atmosphere. versity, Montreál, Que´bec, Canada. 3 To whom requests for reprints should be addressed, at the Institut de Recherches Transient Transfection and Gli Reporter Activity Assay. Mouse cDNAs Cliniques de Montreál, 110 Avenue des Pins, ouest, Montreál, Que´bec, H2W 1R7 Canada. for Gli-1, Gli-2, Gli-3, Ptc, Smo, and suppressor of fused were obtained by Fax: (514) 987-5767; E-mail: [email protected]. reverse transcription of RNA from newborn mouse skin followed by PCR. 4 The abbreviations used are: BCC, basal cell carcinoma; RA, all-trans retinoic acid; Expression plasmids for human Gli-1, Gli-2, and Gli-3 as well as the Gli- RAR, retinoic acid receptor; RXR, retinoid X receptor; CBP, cyclic AMP-responsive element-binding protein-binding protein; Shh, sonic hedgehog; Ptc, patched; Smo, response element reporter plasmid were provided by Dr. C. C. Hui (The smoothened; AP-1, activator protein 1; SMEM, S-minimal essential medium. Hospital for Sick Children, Toronto, Canada). The p300 and CBP expression 5386

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2000 American Association for Cancer Research. RA INHIBITS GLI ACTIVITY vectors were provided by Dr. R. Goodman (Vollum Institute, Portland, OR). All DNAs were purified by Qiagen column chromatography as recommended by the manufacturer. Transient transfections were performed using Lipofect ACE reagent (Life Technologies, Inc.). Briefly, cells were plated in 6-well cluster plates at 4 ϫ 104 cells/well. Transfections consisted of 0.5 ␮gofGli reporter or appropriate control, alone or with expression vectors encoding Gli-1, Gli-2, Gli-3, CBP,orp300 (0.1–1.0 ␮g). DNA mixtures (5 ␮g/transfection; normalized with KSϩ) were combined with 10 ␮l of Lipofect Ace reagent and added to 100 ␮l of serum-free SMEM. This mixture was then added to cells in 1 ml of complete media containing RA or vehicle, as appropriate, and incubated at 37°C overnight. Media were changed daily, and luciferase activity was assessed 48 h after transfection as described previously (20). Results were corrected for protein concentration and are expressed as the mean Ϯ SD from three independent transfections. All experiments were repeated at least three times with comparable results. Northern Blot Analysis. Total RNA (15 ␮g) isolated by Trizol reagent (Life Technologies, Inc.) was size fractionated on a 1% agarose-formaldehyde gel in 4-morpholinepropanesulfonic acid buffer and transferred to a MAGNA Fig. 2. Northern blot analysis of components of the Shh pathway. A Northern blot was prepared using total RNA (15 ␮g) from the different keratinocyte cell lines treated for 48 h Ϫ6 nylon membrane (MSI) according to the manufacturer’s directions. Fragments with carrier (Ϫ)or10 M RA (ϩ). The blot was hybridized with probes generated from used to make probes were isolated by restriction digestion of the appropriate cDNAs (as indicated on the right). Hybridization to 18S rRNA was used as a control. plasmid, followed by purification with Geneclean. Probes were then generated by labeling with [␣-32P]ATP using a labeling kit (Pharmacia). Membranes were hybridized and washed according to the manufacturer’s directions. of RA on the Gli reporter. This could be due to events specific to RAR␥ or to the prevalence of this receptor type (relative to RAR␣)in Results these cell lines (20). To address this issue, the RAR␣␥Ϫ/Ϫ line (which RAR Regulation of Gli Transcriptional Activity in Mouse is devoid of all RARs; Ref. 20) was transfected with the Gli reporter Ϫ6 ␣ ␥ Keratinocyte Cultures. RA (10 M) inhibited Gli reporter activity and various amounts of RAR or RAR expression vectors, and 5–8-fold in wild-type and RAR␣Ϫ/Ϫ cultures. In contrast, reporter reporter activity was assessed in the absence or presence of RA. Both ␣ ␥ activity was either unaffected or only marginally inhibited in RAR and RAR isoforms inhibited reporter activity to similar RAR␥Ϫ/Ϫ cultures, whereas activity in RAR␣␥Ϫ/Ϫ cultures was never degrees in a dose-dependent manner (Fig. 1B). This suggests that the attenuated (Fig. 1A). differential effects of RAR ablation on Gli activity are due to the These results indicate that RAR␥ is the main mediator of the effects relative expression levels of receptor types. Effects of RA on Expression of Constituents of the Shh Signal- ing Pathway. To further investigate the basis for the impact of retinoid treatment on Gli activity, we determined the effect of treatment on the expression of several components of the Shh signaling pathway in wild-type and RAR null lines. In untreated cultures, Gli-2 expression was comparable across all four cell lines, with the excep- tion of reduced levels in the RAR␣ null cell line (Fig. 2). RA reduced Gli-2 levels in both wild-type and RAR␣Ϫ/Ϫ cultures, and this effect was not observed in either RAR␥Ϫ/Ϫ or RAR␣␥Ϫ/Ϫ cell lines. Ptc transcript abundance and RA response across the various cell lines were similar to that of Gli-2. This observation suggests a relationship between expression of these two factors, consistent with previous studies demonstrating regulation of Ptc by Gli. Gli-3 expression was similar in wild-type and RAR␣Ϫ/Ϫ cell lines but was elevated in RAR␥Ϫ/Ϫ and RAR␣␥Ϫ/Ϫ cultures. Gli-3 transcripts were not affected by RA regardless of RAR status. The expression levels of Smo (Fig. 2) and suppressor of fused (data not shown) did not vary significantly among the cell lines. Gli-1 and Shh were not detected. A Role for CBP in RA Repression of Gli. As has been shown for many other transcription factors, such as RARs, Gli-3 transcriptional activity is augmented by the coactivator CBP (18). Whether or not Gli-2 exhibits similar cofactor requirements has not yet been reported. To address the possibility that RA excess might affect Gli function by sequestration of such a common ancillary factor, we tested the effects of exogenous CBP or p300 on Gli reporter activity after RA treatment. CBP or p300 transfection in wild-type cells resulted in a dose- Fig. 1. Effects of RA on Gli activity. A, Gli reporter vector (0.5 ␮g/well) was dependent increase in Gli reporter activity in the absence of RA (Fig. transfected into transformed keratinocyte cell lines, and cells were treated with vehicle (f) 3A). When calculated as fold inhibition, it was found that CBP was Ϫ6 Ⅺ or RA (10 M; ). Luciferase activity was determined from the lysates as described in very effective at reversing the inhibitory effects of RA, with a gain of “Materials and Methods,” and results are the mean Ϯ SD of three independent transfections. B, empty vector or RAR␣1, RAR␣2, RAR␥1, or RAR␥2 expression vectors (0.1 or up to 60% of activity seen in untreated cells (Fig. 3B). Interestingly, 0.5 ␮g/well) were transfected into RAR␣␥ null keratinocytes, and cells were treated as p300 was much less effective in reversing the effects of RA, suggest- described above. f, expression in untreated RAR␣␥ cultures; Ⅺ, expression in cells treated with RA. Samples transfected with RAR expression vectors are denoted by the ing that CBP may be preferred over p300 in this context. This RAR type at the bottom of the figure. difference is not due to lack of expression or function of p300 because 5387

mutual coactivators such as p300 and CBP (17). This mechanism has been described in a number of systems and is thought to underlie at least some of the pharmacological effects of retinoids on tumor growth (16). Interestingly, evidence suggests that similar trans-repression activity is intrinsic to the normal function of the glucocorticoid receptor (21), suggesting that such interactions may also be physio- logically relevant with regard to the RARs. The interaction between AP-1 and RARs predicts that other transcription factors that use p300/CBP, such as Gli-3, should be affected in a similar manner. Consistent with this, we found that wild-type and RAR␣ null cell lines exhibited a profound attenuation of Gli activity by RA. In marked contrast, the RAR␥ null cell line showed minimal effects, and RAR␣␥ null cells were not responsive. Whereas recon- stitution experiments suggest that this effect can be mediated equally by RAR␣ or RAR␥, there appears to be insufficient RAR␣ to elicit an outcome in RAR␥ null cultures. These results correlate well with our previously reported effects of RA on AP-1 (20) and suggest that a common basis underlies the inhibitory effect of retinoid treatment on Gli and AP-1 signaling. Multiple mechanisms have been shown to regulate Gli activity. These include modulation of expression, availability of coactivators, protein kinase A activity, proteolytic processing, and subcellular trafficking (22). It is possible that any or all of these events could be affected by RA treatment. We investigated two of these, coactivator segregation and gene expression, to attempt to explain the effect of RA on Gli activity. Transfection of either CBP or p300 induced basal Gli activity in wild-type keratinocytes to similar degrees, suggesting that, as in other systems, CBP and p300 are functionally similar with regard to tran- Fig. 3. Rescue of RA-mediated repression of Gli activity. Wild-type keratinocytes were transfected with a Gli reporter vector in the absence or presence of various scription through a Gli-responsive promoter. In contrast, CBP was amounts of expression vectors encoding CBP, p300, Gli-1, Gli-2, or Gli-3 (0.1, 0.5, or 1.0 more efficient in attenuating the inhibitory effects of RA. The reversal Ϫ6 ␮g). A, cells were treated with vehicle (f)or10 M RA (Ⅺ), and luciferase activity was by CBP is not surprising, given that both Gli-3 and cubitus interrup- assessed 48 h after transfection as described in “Materials and Methods.” Results are expressed as the percentage of activity relative to untreated wild-type control. B, the tus, the Drosophila homologue of Gli, also use CBP as a coactivator results from A were expressed as the fold Gli activity relative to untreated transfected (18, 19). However, the relative ineffectiveness of p300 in evoking a Ϯ cultures. Results are the mean SD of three independent transfections for both A and B. similar response may be indicative of differential interaction with the The experiment was repeated three times with similar results. RARs. Differential effects manifested by these two highly homolo- gous proteins have been suggested previously for RA-induced differentiation of F9 embryocarcinoma cells (23). untreated cells transfected with the p300 expression vector exhibited Northern blot analysis demonstrated that RA reduced Ptc and Gli-2 a dose-dependent gain of reporter activity comparable to that seen levels. The effects of transfection studies indicate that Gli-2 is a potent with CBP transfection (Fig. 3A). activator of the reporter; hence, this attenuation of Gli-2 expression As described above, RA also has an effect on Gli-2 expression, and may contribute to RA-induced repression of the Gli pathway. Al- this could contribute to the observed inhibition of Gli transcriptional though Gli-1 and Gli-3 were not as potent at affecting basal expres- activity. To assess this, vectors encoding the various Gli members sion levels of the reporter, all three were equipotent at reversing RA were cotransfected with the reporter, and activity was assessed in the repression. However, it is unclear whether all three mediated this absence or presence of RA. Overexpression of Gli-2 resulted in a large effect via a common mechanism. Indeed, to date, only Gli-3 has been (7–8-fold) increase in basal Gli activity, whereas Gli-1 and Gli-3 were shown to functionally associate with CBP. less potent (Fig. 3A). This suggests that Gli-2 may be the main Given that Gli-2 and Ptc are known targets of Shh signaling and are mediator of Gli activity in this system, an observation that is further thus regulated by Gli, it is possible that the down-regulation of Gli-2 supported by our inability to detect Gli-1 and by the low levels of and Ptc by RA is secondary to a reduction in Gli signaling. In such a Gli-3 in wild-type cultures. However, all three Gli members were very model, CBP sequestration would lead to decreased Gli activity, which potent at reversing the inhibitory effect of RA on Gli activity, with a in turn would result in decreased expression of Gli target genes, recovery of 70–80% of expression relative to the untreated control including Gli-2 and Ptc. The ability of CBP to reverse this effect is (Fig. 3B). consistent with such a model. Although the rescue effect was very pronounced with CBP and with Shh signaling is essential for normal skin development and prolif- different Gli members, repression of Gli activity was never com- eration. In our model, Shh and Gli-1 transcripts were undetectable, pletely abolished. This observation suggests that both titration of indicating that the observed basal activity is not dependent on Shh. In limiting cofactors and inhibition of Gli family member expression by contrast, BCCs often overexpress Gli-1 due to inappropriate activation RA may act in concert to attenuate Gli activity in these transformants. of the Shh pathway (24). Although there are a number of contradictory Discussion studies, BCC does not always respond positively to retinoid therapy (25). This may be due to the genetic basis of these tumors because The RARs can trans-repress the activity of certain transcription Gli-1 does not require CBP for efficient transcriptional activity and factors, such as AP-1, through a mechanism implicating titration of may not be affected by RA. It remains to be seen whether the present 5388

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2000 American Association for Cancer Research. RA INHIBITS GLI ACTIVITY observations are unique to this particular model system or whether RA 11. Lotan, R. Retinoids and chemoprevention of aerodigestive tract cancers. Cancer can affect Shh signaling in BCCs. RA has been shown to impact Gli Metastasis Rev., 16: 349–356, 1997. 12. Lotan, R. Suppression of squamous cell carcinoma growth and differentiation by and Shh expression in Xenopus and Shh expression in mouse embryos retinoids. Cancer Res., 54 (Suppl.): 1987s–1990s, 1994. (26, 27). Moreover, we have found that RA treatment also alters Gli-1 13. Leid, M., Kastner, P., Durand, B., Krust, A., Leroy, P., Lyons, R., Mendelsohn, C., expression in late gastrulation mouse embryos (our preliminary re- Nagpal, S., Nakshatri, H., Reibel, C., et al. Retinoic acid signal transduction pathways. Ann. N. Y. Acad. Sci., 684: 19–34, 1993. sults). These observations further support our findings suggesting 14. Fisher, G. J., and Voorhees, J. J. Molecular mechanisms of retinoid actions in skin. interplay between these important signaling pathways. FASEB J., 10: 1002–1013, 1996. 15. De Luca, L. M., Darwiche, N., Celli, G., Kosa, K., Jones, C., Ross, S., and Chen, L. C. Acknowledgments Vitamin A in epithelial differentiation and skin carcinogenesis. Nutr. Rev., 52: S45–S52, 1994. 16. Huang, C., Ma, W. Y., Dawson, M. I., Rincon, M., Flavell, R. A., and Dong, Z. We thank P. Chambon for the RAR null mice, Drs. C. C. Hui and R. Blocking activator protein-1 activity, but not activating retinoic acid response ele- Goodman for vectors, and members of the laboratory for support and sugges- ment, is required for the antitumor promotion effect of retinoic acid. Proc. Natl. Acad. tions. Sci. USA, 94: 5826–5830, 1997. 17. Kamei, Y., Xu, L., Heinzel, T., Torchia, J., Kurokawa, R., Gloss, B., Lin, S. C., References Heyman, R. A., Rose, D. W., Glass, C. K., and Rosenfeld, M. G. A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. 1. Xie, J. W., Murone, M., Luoh, S. M., Ryan, A., Gu, Q. M., Zhang, C. H., Bonifas, Cell, 85: 403–414, 1996. J. M., Lam, C. W., Hynes, M., Goddard, A., Rosenthal, A., Epstein, E. H., and 18. Akimaru, H., Chen, Y., Dai, P., Hou, D. X., Nonaka, M., Smolik, S. M., Armstrong, Desauvage, F. J. Activating smoothened mutations in sporadic basal-cell carcinoma. S., Goodman, R. H., and Ishii, S. Drosophila CBP is a co-activator of cubitus Nature (Lond.), 391: 90–92, 1998. interruptus in hedgehog signalling. Nature (Lond.), 386: 735–738, 1997. 2. Dahmane, N., Lee, J., Robins, P., Heller, P., and Altaba, A. R. I. Activation of the 19. Dai, P., Akimaru, H., Tanaka, Y., Maekawa, T., Nakafuku, M., and Ishii, S. Sonic transcription factor Gli-1 and the sonic hedgehog signalling pathway in skin tumours. hedgehog-induced activation of the Gli1 promoter is mediated by GLI3. J. Biol. Nature (Lond.), 389: 876–881, 1997. Chem., 274: 8143–8152, 1999. 3. Fan, H. R., Oro, A. E., Scott, M. P., and Khavari, P. A. Induction of basal cell 20. Goyette, P., Chen, C. F., Wang, W., Seguin, F., and Lohnes, D. Characterization of carcinoma features in transgenic human skin expressing sonic hedgehog. Nat. Med., retinoic acid receptor (RAR)-deficient keratinocytes. J. Biol. Chem., 275: 16497– 3: 788–792, 1997. 16505, 2000. 4. Johnson, R. L., Rothman, A. L., Xie, J., Goodrich, L. V., Bare, J. W., Bonifas, J. M., 21. Reichardt, H. M., Kaestner, K. H., Tuckermann, J., Kretz, O., Wessely, O., Bock, R., Quinn, A. G., Myers, R. M., Cox, D. R., Epstein, E. H. J., and Scott, M. P. Human Gass, P., Schmid, W., Herrlich, P., Angel, P., and Schutz, G. DNA binding of the homolog of patched, a candidate gene for the basal cell nevus syndrome. Science glucocorticoid receptor is not essential for survival. Cell, 93: 531–541, 1998. (Washington DC), 272: 1668–1671, 1996. 22. Hahn, H., Wojnowski, L., Miller, G., and Zimmer, A. The patched signaling pathway 5. Goodrich, L. V., Milenkovic, L., Higgins, K. M., and Scott, M. P. Altered neural cell fates and medulloblastoma in mouse patched mutants. Science (Washington DC), in tumorigenesis and development: lessons from animal models. J. Mol. Med., 77: 277: 1109–1113, 1997. 459–468, 1999. 6. Kinzler, K. W., Bigner, S. H., Bigner, D. D., Trent, J. M., Law, M. L., O’Brien, S. J., 23. Kawasaki, H., Eckner, R., Yao, T. P., Taira, K., Chiu, R., Livingston, D. M., and Wong, A. J., and Vogelstein, B. Identification of an amplified, highly expressed gene Yokoyama, K. K. Distinct roles of the co-activators p300 and CBP in retinoic in a human glioma. Science (Washington DC), 236: 70–73, 1987. acid-induced F9 cell differentiation. Nature (Lond.), 393: 284–289, 1998. 7. Roberts, W. M., Douglass, E. C., Peiper, S. C., Houghton, P. J., and Look, A. T. 24. Altaba, A. R. I. Gli proteins and Hedgehog signaling: development and cancer. Amplification of the gli gene in childhood sarcomas. Cancer Res., 49: 5407–5413, Trends Genet., 15: 418–425, 1999. 1989. 25. Evans, T. R., and Kaye, S. B. Retinoids: present role and future potential. Br. 8. Chambon, P. A decade of molecular biology of retinoic acid receptors. FASEB J., 10: J. Cancer, 80: 1–8, 1999. 940–954, 1996. 26. Wasiak, S., and Lohnes, D. Retinoic acid affects left-right patterning. Dev. Biol., 215: 9. De Luca, L. M. Retinoids and their receptors in differentiation, embryogenesis, and 332–342, 1999. neoplasia. FASEB J., 5: 2924–2933, 1991. 27. Franco, P. G., Paganelli, A. R., Lopez, S. L., and Carrasco, A. E. Functional 10. Singh, D. K., and Lippman, S. M. Cancer chemoprevention. I. Retinoids and caro- association of retinoic acid and hedgehog signaling in Xenopus primary neurogenesis. tenoids and other classic antioxidants. Oncology (Basel), 12: 1643–1660, 1998. Development (Camb.), 126: 4257–4265, 1999.

5389

Philippe Goyette, Deborah Allan, Pascal Peschard, et al.

Cancer Res 2000;60:5386-5389.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/60/19/5386

Cited articles This article cites 27 articles, 9 of which you can access for free at: http://cancerres.aacrjournals.org/content/60/19/5386.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/60/19/5386.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/60/19/5386. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.